Since January 2020 Elsevier has created a COVID-19 resource centre with

free information in English and Mandarin on the novel coronavirus COVID-

19. The COVID-19 resource centre is hosted on Elsevier Connect, the
company's public news and information website.

Elsevier hereby grants permission to make all its COVID-19-related

research that is available on the COVID-19 resource centre - including this
research content - immediately available in PubMed Central and other
publicly funded repositories, such as the WHO COVID database with rights
for unrestricted research re-use and analyses in any form or by any means
with acknowledgement of the original source. These permissions are
granted for free by Elsevier for as long as the COVID-19 resource centre
remains active.
 American Journal of Infection Control 45 (2017) 1308-11

Contents lists available at ScienceDirect

American Journal of Infection Control American Journal of

Infection Control

j o u r n a l h o m e p a g e : w w w. a j i c j o u r n a l . o r g

Major Article

Contact tracing with a real-time location system: A case study of

increasing relative effectiveness in an emergency department
Thomas R. Hellmich MD a,b, Casey M. Clements MD, PhD a,b, Nibras El-Sherif MBBS b,c,
Kalyan S. Pasupathy PhD b,c, David M. Nestler MD, MS a,b, Andy Boggust MD a,
Vickie K. Ernste DNP, RN a, Gomathi Marisamy BS d, Kyle R. Koenig BS c,
M. Susan Hallbeck PhD, CPE, PE b,c,d,*
a Department of Emergency Medicine, Mayo Clinic, Rochester, MN
b Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, Mayo Clinic, Rochester, MN
c Department of Health Sciences Research, Mayo Clinic, Rochester, MN
d
Departments of Surgery and Critical Care Systems, Mayo Clinic, Rochester, MN

Key Words: Background: Contact tracing is the systematic method of identifying individuals potentially exposed to
Infectious disease infectious diseases. Electronic medical record (EMR) use for contact tracing is time-consuming and may
Pertussis miss exposed individuals. Real-time location systems (RTLSs) may improve contact identification. There-
fore, the relative effectiveness of these 2 contact tracing methodologies were evaluated.
Methods: During a pertussis outbreak in the United States, a retrospective case study was conducted
between June 14 and August 31, 2016, to identify the contacts of confirmed pertussis cases, using EMR
and RTLS data in the emergency department of a tertiary care medical center. Descriptive statistics and
a paired t test (α = 0.05) were performed to compare contacts identified by EMR versus RTLS, as was cor-
relation between pertussis patient length of stay and the number of potential contacts.
Results: Nine cases of pertussis presented to the emergency department during the identified time period.
RTLS doubled the potential exposure list (P < .01). Length of stay had significant positive correlation with
contacts identified by RTLS (ρ = 0.79; P = .01) but not with EMR (ρ = 0.43; P = .25).
Conclusions: RTLS doubled the potential pertussis exposures beyond EMR-based contact identification.
Thus, RTLS may be a valuable addition to the practice of contact tracing and infectious disease monitoring.
© 2017 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier
Inc. All rights reserved.

Contact tracing is a critical strategy required for timely preven-
tion and control of infectious disease outbreaks. 1,2 However,
conventional contact tracing methods are time-consuming and can
miss a significant number of potential exposures.3,4 Medical record
review and staff interview methods fail to capture all potentially
exposed individuals because of incomplete or missing documen-
tation and recall bias.5,6 Failure to identify persons in contact with
infected patients increases the risk of transmission, placing many
* Address correspondence to M. Susan Hallbeck, PhD, CPE, PE, Robert D. and Patricia
E. Kern Center for the Science of Health Care Delivery, Mayo Clinic, Division of Health
Care Policy and Research, Department of Health Sciences Research, Mayo Clinic, 200
First St SW, Rochester, MN 55905.
E-mail address: hallbeck.susan@mayo.edu (M.S. Hallbeck).
Supported by funds from the Mayo Clinic Robert D. and Patricia E. Kern Center
for the Science of Health Care Delivery and contributions from the Mayo Clinic ED-
Clinical Engineering Learning Lab.
Conflicts of interest: None to report.
health care workers and vulnerable patients (eg, infants and
comorbid patients) at risk.7 This transmission may cause signifi-
cant health complications, especially for vulnerable patients, and
increase morbidity and mortality.7 Failure to adequately trace in-
fectious disease contacts disrupts routine health care services8 and
results in substantial cost for health care systems.9,10 Current contact
tracing methodologies for contagious diseases are imperfect, and
new technological interventions should be investigated to identi-
fy close contacts in a timely, efficient, and exhaustive manner to
prevent subsequent transmission to other patients and health care
workers for effective outbreak management.
Advances in technology have made tracking individuals possi-
ble and increasingly affordable using several types of real-time
location system (RTLS). One such RTLS uses radiofrequency iden-
tification (RFID) tracking. RFID has been used in a variety of settings
such as schools11,12 and academic conferences13,14 to facilitate and
accelerate the process of understanding face-to-face contact, human
interactions, and social networks accurately and efficiently within

0196-6553/© 2017 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
https://doi.org/10.1016/j.ajic.2017.08.014
 T.R. Hellmich et al. / American Journal of Infection Control 45 (2017) 1308-11 1309

Table 1
Nine pertussis cases and potential exposure identified by electronic medical record (EMR) and real-time location system (RTLS)

Emergency No. by EMR No. by RTLS Total Increase

department length No. by EMR No. by RTLS (but not (but not unique above EMR
Case No. Age, y of stay, min review review RTLS) EMR) contacts only, %
1 1.6 91 4 8 2 6 10 150.0
2 18 152 5 9 1 5 10 100.0
3 11 144 7 9 1 3 10 42.9
4 2.1 68 5 8 2 5 10 100.0
5 17 121 5 5 1 1 6 20.0
6 17 62 3 3 0 0 3 0
7 9 288 5 10 2 7 12 140.0
8 2.8 287 5 14 2 11 16 220.0
9 8 294 6 11 2 7 13 116.7
Total 45 77 13 45 90
Average 9.6 94 5.0 8.6 1.4 5.0 10.0 100

a very short time. In health care settings, RFID has been used to study
the role of patient-health care worker networks and face-to-face
contact frequency and duration in the disease transmission process15
and to identify the most and least connected health care workers
with patients.16
To date, no study has compared RTLS with the standard meth-
odology of chart review for contact tracing. In this case study, we
describe a relative effectiveness comparison of contact tracing
between current state (chart review) and RTLS tracking for con-
firmed pertussis cases in an emergency department (ED).
MATERIALS AND METHODS
Equipment
Our RTLS system (Quake Global, Inc, San Diego, CA) utilizes 194
in-ceiling, passive RFID readers with 734 antennas, covering 212 lo-
cations in a 54,450 sq ft ED and radiology (supporting ED services)
department space. The RTLS, driven by this RFID infrastructure, was
in operation and reliable for 6 months before data collection. Core
ED staff wore RFID-enabled employee identification tags. In addi-
tion to physicians and nurses, staff including respiratory therapists
and registration personnel had RFID-enabled badges. However, some
ancillary staff serving the ED (eg, medical students and radiology
technicians) did not have RFID-enabled identification tags during
the time of this study.
Study protocol
During an outbreak of pertussis in the Midwest region of the
United States, a retrospective case study was conducted to identi-
fy possible contacts of 9 confirmed, successive pertussis cases
between June 14 and August 31, 2016. Our study was performed
in the ED of a large tertiary medical center with an annual volume
of approximately 74,000 patient encounters. We used both tradi-
tional electronic medical record (EMR)-based contact identification
and RTLS-generated data. All pertussis cases were diagnosed 1-2
days after the ED visit and reported to institutional infection control
services. None of these 9 patients were admitted to the hospital.
Per existing contact tracing protocol, a list of possible expo-
sures for each pertussis patient was generated after review of the
EMR. Nurse leadership in coordination with hospital infection control,
using current Centers for Disease Control and Prevention guide-
lines, conducted this review. Possible disease exposure included any
health care worker whose role placed them in face-to-face contact
with the index patient in the exam room or triage area. The EMR
identified any health care workers who documented their interac-
tions with patients.
Subsequently, a list of possible pertussis exposures was gener-
ated using a novel RTLS event mapping program that identified
patients and staff, their colocations, and movements over time and
space. It took <5 minutes to generate each of these RTLS data queries.
Because pertussis is spread through droplet respiratory secre-
tions, possible exposures were limited to colocation in ED exam
rooms or triage areas where droplet exposures are most likely.
Hallway and waiting room colocation were excluded from the anal-
ysis because they were not considered likely to result in face-to-
face exposure. All possible exposures, identified through RTLS or
EMR review, were offered standard pertussis exposure prophylaxis.
Analysis
We compared the number of possible exposures identified by
EMR, the number of possible exposures identified by RTLS, and the
roles of the health care staff that came in contact with confirmed
pertussis cases. Descriptive statistics and paired t tests were used
to compare the number of contacts and the role of those identi-
fied by EMR and RTLS, with α = 0.05. In addition, a Pearson product-
moment correlation between the length of stay (LOS) and the
number of contacts identified by each methodology was performed.
RESULTS
Nine patients eventually diagnosed with pertussis presented to
the ED during our identified time period, with patient ages ranging
from 1.6-18 years (mean, 9.6 years). The average LOS was 94 minutes
(range, 62 to 294 minutes; standard deviation = 96.6).
When combining all 9 cases, EMR review (alone) identified 45
potential contacts (Table 1 and Fig 1). Thirteen health care workers
were identified in the EMR but not identified by RTLS, including 2
physicians, 6 triage nurses, 1 discharge nurse, 1 scribe, 1 urology
technician, 1 radiology technician, and 1 medical student. RTLS alone
identified 77 contacts, of whom 45 were additional new contacts
not identified from the EMR. Table 2 shows the roles of the health
care staff identified by RTLS but not by EMR review. RTLS identi-
fied twice as many possible contact cases as those identified by EMR
review (P < .01) increasing the number of potential contacts above
those identified by the EMR from an average of 5 contacts per case
to an average of 10 contacts per case (Fig 1 and Table 1).
DISCUSSION
In this study, RTLS doubled the number of potential pertussis
exposures identified beyond the conventional methods of EMR-
based contact identification, suggesting that RTLS may be more
effective than traditional EMR review alone. RTLS-generated data
1310 T.R. Hellmich et al. / American Journal of Infection Control 45 (2017) 1308-11

Table 2
Health care staff additionally identified as potentially exposed by real-time location system

Role Case No. Total, n Total additional staff identified, %

1 2 3 4 5 6 7 8 9
Physician 1 1 2 4.4
Nurse 2 3 1 2 2 4 2 16 35.6
Pharmacist 1 1 2.2
Respiratory care 1 1 1 3 6.7
Care team assistant 2 2 1 1 3 9 20.0
Registration 1 1 1 2 1 6 13.3
Phlebotomist 2 2 4.4
Child life specialist 1 1 1 2 1 6 13.3
Total 6 5 3 5 1 0 7 11 7 45 100

NOTE. Case numbers 7, 8, and 9 had the longest length of stay (4:48, 4:47, and 4:90 min, respectively) and also had the largest number of additionally identified staff by
RTLS (7, 11, and 7, respectively) (Table 2). Length of stay had significant positive correlation with the total unique contacts (ρ = 0.78; P = .01). When further examined by
identification method, length of stay had significant positive correlation with the total contact number identified by RTLS (ρ = 0.79; P = .01) but length of stay was not sig-
nificantly correlated with the total contact number identified by EMR review (ρ = 0.43; P = .25).

Fig 1. Electronic medical record (EMR) versus real-time location system (RTLS) in
identifying pertussis contacts.
can be a valuable adjunct to identify additional potential disease
exposures, especially health care staff who are not principal care-
takers and/or those in supporting ED roles.
RTLS appears to be of greater benefit when patients have longer
ED LOS. We showed statistically significant positive correlation with
ED LOS and RTLS-identified contacts, but not with EMR-identified
contacts. The literature has shown that the LOS of contagious pa-
tients in hospital settings increases the potential for nosocomial
spread of contagious diseases,17,18 but no study has looked into the
correlation between LOS and number of identified contacts using
different contact tracing methodologies. Our finding may be due to
the fact that patients with longer LOS interact with more health care
providers during their stay.
In addition to the increased identification of exposed health care
staff, RTLS also saves time and resources in identifying potential ex-
posures due to rapidity and can quantify exposure duration. In this
study, RTLS data query took, on average, <5 minutes to generate a
list of possible exposures and unlike chart review gave duration of
exposure. Time to complete standard chart review was not specif-
ically measured, but was estimated to be 30-60 minutes per patient
EMR. Other studies have found that the use of chart review in contact
tracing is time-consuming.3 The objective data provided by RTLS
reduced the need to further verify and interview staff docu-
mented within the chart to identify additional potential exposures.
In addition, RTLS may also reduce the effect of recall bias and need
to contact health care staff, who must also rely on memory, in an
effort to identify potential exposures.
The RTLS system failed to identify 13 health care workers who
were documented in the medical record. Four were representa-
tives of staff categories that had not been equipped with RFID-
enabled badges, and 6 were triage nurses who provided care before
the rooming process and their RTLS antenna was moved to in-
crease probabilities of registering in the future. Two physicians were
missed, 1 whose badge had an RFID chip that was not being read
correctly by the RTLS, and the other was a resident physician who
was not wearing the correct RFID-enabled badge. There was not a
clear explanation why 1 discharge nurse was not picked up in the
query. Further expansion of RIFD-enabled badges and analysis of
system outliers is needed to ensure overall RTLS system integrity.
Although the cost of the RTLS system can be justifiable in large
urban EDs with diverse patient populations, some might argue
against such high cost in small community EDs that are less likely
to develop major infectious disease outbreaks. However, diseases
like pertussis are increasingly reemerging in cycles every 3-5 years
despite the high rate of vaccination,19 and lack of vaccination is an
increasing concern.20 In other studies, nosocomial pertussis out-
breaks resulted in substantial cost to hospitals, even when the
number of pertussis cases was low.9,10 Also, contact tracing is only
1 method to justify the cost of RTLS, and some smaller communi-
ty EDs may also attain a reasonable return on investment using
metrics such as reducing asset shrinkage.21
Limitations to our study include, but are not limited to, a small
sample size. We tested only 1 ED, and focused on 1 specific disease,
which makes it less generalizable. Also, this study did not include
the mitigating effect of personal protective equipment use by health
care staff members for exposure to pertussis-diagnosed patients, and
did not take the cost of using each contact tracing method into
account. Further multicenter comparative studies of infectious dis-
eases using contact tracing between the RTLS and EMR, with larger
sample sizes and with focus on more than 1 disease, are needed
to confirm our findings and cost-effectiveness and allow for more
generalizability.
CONCLUSIONS
RTLS identified more potential contacts with pertussis-diagnosed
patients than EMR-based chart review. Although RTLS is not mature
enough a technology to entirely replace standard chart review, this
case study illustrates how an evolving technology can accelerate the
process of active screening and facilitate timely cessation in the chain
of transmission of infectious diseases. RTLS can be an important tool
for infection control programs, especially in tertiary care medical
systems with higher acuity and complex, vulnerable patients. Such
hospitals are relied upon to care for patients with rare, high-risk,
contagious diseases like measles, tuberculosis, and emerging in-
fections such as Ebola virus disease, severe acute respiratory
syndrome, or Middle East respiratory syndrome. Notably, several
high-risk emerging infections are transmitted by airborne spread
 T.R. Hellmich et al. / American Journal of Infection Control 45 (2017) 1308-11
with greater contagious risk to staff and other patients, and if dis-
charged home, to the public at large. We believe that RTLS may show
a more substantial benefit in these cases, because brief exposures
at a distance are clinically significant and difficult to thoroughly eval-
uate by traditional means, whereas event mapping using RTLS is
customizable to pick up such brief exposures.
References
1. Browne C, Gulbudak H, Webb G. Modeling contact tracing in outbreaks with
application to Ebola. J Theor Biol 2015;384:33-49.
2. Begun M, Newall AT, Marks GB, Wood JG. Contact tracing of tuberculosis: a
systematic review of transmission modelling studies. PLoS One 2013;8:e72470.
3. Dixon MG, Taylor MM, Dee J, Hakim A, Cantey P, Lim T, et al. Contact tracing
activities during the Ebola virus disease epidemic in Kindia and Faranah, Guinea,
2014. Emerg Infect Dis 2015;21:2022-8.
4. England DO, Currie MJ, Bowden FJ. An audit of contact tracing for cases of
chlamydia in the Australian Capital Territory. Sex Health 2005;2:255-8.
5. Bryant KA, Humbaugh K, Brothers K, Wright J, Pascual FB, Moran J, et al. Measures
to control an outbreak of pertussis in a neonatal intermediate care nursery after
exposure to a healthcare worker. Infect Control Hosp Epidemiol 2006;27:541-5.
6. Daskalaki I, Hennessey P, Hubler R, Long SS. Resource consumption in the
infection control management of pertussis exposure among healthcare workers
in pediatrics. Infect Control Hosp Epidemiol 2007;28:412-7.
7. O’Brien JA, Caro JJ. Hospitalization for pertussis: profiles and case costs by age.
BMC Infect Dis 2005;5:57.
8. Ward A, Caro J, Bassinet L, Housset B, O’Brien JA, Guiso N. Health and economic
consequences of an outbreak of pertussis among healthcare workers in a hospital
in France. Infect Control Hosp Epidemiol 2005;26:288-92.
9. Baggett HC, Duchin JS, Shelton W, Zerr DM, Heath J, Ortega-Sanchez IR, et al.
Two nosocomial pertussis outbreaks and their associated costs—King County,
Washington, 2004. Infect Control Hosp Epidemiol 2007;28:537-43.
1311
10. Zivna I, Bergin D, Casavant J, Fontecchio S, Nelson S, Kelley A, et al. Impact of
Bordetella pertussis exposures on a Massachusetts tertiary care medical system.
Infect Control Hosp Epidemiol 2007;28:708-12.
11. Salathe M, Kazandjieva M, Lee JW, Levis P, Feldman MW, Jones JH. A high-
resolution human contact network for infectious disease transmission. Proc Natl
Acad Sci USA 2010;107:22020-5.
12. Stehle J, Voirin N, Barrat A, Cattuto C, Isella L, Pinton JF, et al. High-resolution
measurements of face-to-face contact patterns in a primary school. PLoS ONE
2011;6:e23176.
13. Cattuto C, Van den Broeck W, Barrat A, Colizza V, Pinton JF, Vespignani A.
Dynamics of person-to-person interactions from distributed RFID sensor
networks. PLoS ONE 2010;5:e11596.
14. Isella L, Romano M, Barrat A, Cattuto C, Colizza V, Van den Broeck W, et al. Close
encounters in a pediatric ward: measuring face-to-face proximity and mixing
patterns with wearable sensors. PLoS ONE 2011;6:e17144.
15. Gundlapalli A, Ma X, Benuzillo J, Pettey W, Greenberg R, Hales J, et al. Social
network analyses of patient-healthcare worker interactions: implications for
disease transmission. AMIA Ann Symp Proc 2009;2009:213-7.
16. Hornbeck T, Naylor D, Segre AM, Thomas G, Herman T, Polgreen PM. Using sensor
networks to study the effect of peripatetic healthcare workers on the spread of
hospital-associated infections. J Infect Dis 2012;206:1549-57.
17. Vincent J, Bihari DJ, Suter PM, Bruining HA, White J, Nicolas-Chanoin MH et al.
The prevalence of nosocomial infection in intensive care units in Europe: results
of the European Prevalence of Infection in Intensive Care (EPIC) study. JAMA
1995;274:639-44.
18. Cohen B, Hyman S, Rosenberg L, Larson E. Frequency of patient contact with
health care personnel and visitors: implications for infection prevention. Jt Comm
J Qual Patient Saf 2012;38:560-5.
19. Bouchez V, Guiso N. Bordetella pertussis, B. parapertussis, vaccines and cycles of
whooping cough. Pathog Dis 2015;73:pii: ftv055.
20. Phadke VK, Bednarczyk RA, Salmon DA, Omer SB. Association between vaccine
refusal and vaccine-preventable diseases in the United States: a review of measles
and pertussis. JAMA 2016;315:1149-58.
21. Rousek JB, Pasupathy K, Gannon D, Hallbeck S. Asset management in healthcare:
evaluation of RFID. IIE Trans Healthc Syst Eng 2014;4:144-55.